Project Summary/Abstract This competing renewal in its 30th year continues a strong record of studies of DNA replication, in particular the architecture of moving forks. The unique approach couples electron microscopic (EM) examination of DNA- protein complexes, biochemical analysis, and the use of genetic mutants. Previous work supported by this grant provided the first direct proof that the lagging strand at a fork loops back (trombone model) and that there are 2 (and sometimes 3) polymerase molecules present. DNA replication in human and animal mitochondria (mt) is central to aging and DNA damage. Recent work by Jacobs and others has revealed that human mtDNA may replicate via modes not previously thought to occur, including the generation of recombinational networks. In deciphering these mechanisms, the EM approaches employed will be critical in establishing the architecture of the replication intermediates. These studies will continue as collaborations with Howard Jacobs and Laurie Kaguni's labs and will focus on the Drosophila mtDNA system where powerful genetic tools are available. Comparative studies of Drosophila and human mt. single strand (ss) binding proteins and their functional variants will be examined in their binding to ssDNA. In addition to examining purified mtDNA, mtDNA-protein complexes from human and Drosophila cells will be examined by EM. Work in progress on the novel rolling circle replication of nematode mtDNA will be continued. In order to understand basic modes of replication in human cells, animal viruses are powerful tools and the Herpes Simplex virus type 1 has provided an excellent model system due to the simple set of 6 replication factors used and that three modes of replication: rolling circle theta, and recombination-coupled replication may be employed in vivo. All 6 HSV-1 replication proteins have been highly purified in this laboratory and using them, a recent groundbreaking advance has been the reconstitution in vitro of robust rolling circle replication which generates tails of >25 kb from circular ?X174 templates. At later incubation times the reactions generate recombinational DNA networks, providing the first model system for studying this poorly understood replication mode. These advances set the stage for rapid progress bolstered by collaboration with Sandra Weller's group who will provide a large number of mutants in the proteins. The studies will focus on understanding the mechanism of all three replication modes and whether there is a lagging strand loop. The number of polymerase and helicase/primase molecules at a moving fork will be determined using a tool developed in the last renewal, nano-scale bipointers. A new EM method (cryo-shadowing) will be applied to achieve better structural resolution. These experiments will test our hypothesis that looping of the lagging strand and the general features of the fork architecture are broadly shared across widely different systems.